© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE Local
    5. 6-RES_SCL-AJT
    6. 3865
    Prev
    Next
    INDEX
    Explanations

    specific skills or abilities described in a technical or professional context

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    •Layer 6 UMAP region: Mixed in middle - local

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    447
    +0.23
    1.2%
    575
    +0.15
    0.8%
    288
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    285
    +0.23
    0.09
    393
    +0.15
    0.26
    283
    +0.12
    0.02
    Negative Logits
    romeda
    -0.70
     Manz
    -0.62
     Wilhelm
    -0.61
     Reloaded
    -0.61
    idays
    -0.60
    clipse
    -0.60
     Pall
    -0.59
    choes
    -0.57
     Amen
    -0.57
     Claus
    -0.57
    POSITIVE LOGITS
    cape
    0.89
    anooga
    0.84
    learn
    0.81
     skills
    0.78
     training
    0.78
     proficiency
    0.76
    chery
    0.76
    poke
    0.75
    linger
    0.74
    cius
    0.72
    Activations Density 11.392%

    No Known Activations